Aromatase inhibitors (AIs) have been rapidly replacing tamoxifen (TAM) as first-line adjuvant hormonal therapy for postmenopausal women diagnosed with hormone receptor (HR)-positive, early-stage breast cancer. The profound estrogen depletion triggered by AIs is responsible for improved outcomes compared to TAM, yet AI therapy can negatively impact bone health, elevating the already high risk of fractures in postmenopausal women. Osteoporotic fractures at older age can result in markedly increased mortality, poor quality of life, and staggering healthcare costs. Despite these outcomes, risk factors for fracture specific to postmenopausal breast cancer patients taking AIs remain surprisingly understudied. To date, no validated tools exist for fracture risk assessment specific to postmenopausal women prior to initiation of AI therapy. This application requests to conduct for the first time a prospective study on bone health in 2,062 postmenopausal breast cancer patients who received AI therapy in the Pathways Study, a prospective cohort study of breast cancer prognosis in the Kaiser Permanente Northern California (KPNC) Medical Care Program, enrolled from 2006-2013 and followed through 2016. The establishment of Pathways was concurrent with AIs widely replacing TAM as hormonal therapy for postmenopausal patients. By leveraging a rich body of epidemiologic, clinical and pharmacy data linked with high-quality biospecimens in Pathways, we have a unique opportunity to conduct one of the first in-depth studies of AI-associated fractures in breast cancer patients. Among postmenopausal women who received AI therapy for early-stage, HR-positive breast cancer, we will investigate the risk of fractures (primary outcome) and osteoporosis (secondary outcome) in association with 1) modifiable lifestyle factors such as physical activity, diet, vitamin D and calcium supplement use, smoking, and alcohol consumption; 2) germline genetic variations in estrogen and bone metabolism pathways with validation of findings using genome-wide assays; and 3) the associations of serum biomarkers, including BAP for bone formation and TRAP5b for resorption, six key regulatory cytokines (RANKL, OPG, IL1, IL6, TNF?, CSF), and 25-hydroxyvitamin D. Finally, a prediction model for fracture risk in postmenopausal breast cancer patients on AI therapy will be developed based upon lifestyle factors, genetic variations, and serum biomarkers and compared with models intended for the general healthy population. Over 2.5 million women with breast cancer live in the U.S. today, with an estimated 230,000 newly diagnosed cases in 2011. An excess of 13,000 fractures per year has been estimated among postmenopausal survivors compared to their healthy counterparts. Therefore, understanding the health effects of AI therapy on risk of skeletal outcomes is of great public health importance. In the Pathways Study, we now have an exceptional opportunity to address an important gap in breast cancer survivorship research, and to reduce the burden of AI-induced osteoporotic fractures in a real-world clinical setting.